Runway Act-Two

Enhanced Performance Capture

Runway Act-Two captures every motion aspect of a character in a driving performance video with incredible accuracy. Whether it's head, face, body, or hand movements, the AI model can render the smallest nuances or complex series of movements in lifelike detail for truly realistic and immersive storytelling.

Multimodal Inputs Supported

Runway Act-Two’s multimodal capabilities let users choose between animating a character in a still image or transferring gestures to a character in a video. With both input options available, the AI model offers more flexible support in leveraging real human performances to craft compelling narrative sequences.

Configurable Gesture Control

With Runway Act-Two, users can decide to transfer body movements from the performance video to the character image or not. If the gesture control is off, the AI model focuses on facial and environmental motion. The only exception is with character video inputs, where the gesture control setting remains unavailable.

Adjustable Facial Expressions

Users can set their desired level of facial expressiveness with Runway Act-Two. The lower the configured value, the less facial expression the characters will have, which means greater character consistency. A higher value results in more expressiveness, but this could potentially contribute to visual artefacts in the final scene.

Adaptable Scene Motion

Depending on the input, Runway Act-Two automatically adjusts the camera and environmental motion control to produce the highest quality results. With image inputs, the AI model implements intuitive changes that best suit the animated character, but for video inputs, the original camera and scene movement remain the same.

Broad Video Applications

Runway Act-Two’s advanced motion-capture capabilities make it well-suited to a broad range of creative visual projects that require lifelike character motion renders. With this AI model, users can produce interactive digital guides, product model scenes, stylized tutorials, narrative game scenes, animation sequences, etc.